Conversation

Thorsten Leemhuis (acct. 1/4)

Edited 1 year ago

6.6 just became a (aka ) .

The 's stable team (Sasha Levin and @gregkh) just made this official by adding it to https://www.kernel.org/category/releases.html with the following change: https://git.kernel.org/pub/scm/docs/kernel/website.git/commit/?id=d3c85f300d9214949efae275e519f30cce155cca

Projected EOL is December 2026.

5
5
2

@kernellogger @gregkh Any plans to reinstate Tuz for 6.6.6 ;)?

1
0
0

@kernellogger @gregkh wouldn't have expected 6.6 to end up as an LTS given the large core changes with the Scheduler, but I am glad to see it happen, never the less.

1
0
0

2/ And in case anyone wonders:

This means that next years longterm will be[1] 6.12, unless something odd happens.

[1] because (a) the last release of a year is almost always declared a longterm kernel these days (b) new mainline versions these days are almost always released every 63 (in roughly about 60% of the cases) or 70 (~40%) days

2
0
1

3/ for details on claim (b) see the chart below which you can also find here: https://docs.google.com/spreadsheets/d/1_yH7lFmZxAoSWrtsd8tGu3befG4zIcMnytB1ml4pQQM/edit#gid=1170907286&range=A129

But please don't check the percentages given, they are just a while guess from the data, I did not check for real; but for the calc in the previous toot that's good enough I'd say.

1
0
0

@mxk @gregkh

that's because you assume code changes have anything to do with the decition. But they don't. The relevant factor is "time of year"; see the reply to the initial tooth for details.

1
0
0

@nik

don't put stupid ideas in @gregkh's head! πŸ˜‚ 😬 πŸ™ƒ

0
0
0

@kernellogger I think I remember @gregkh explicitly saying, that he choose certain versions over others als LTS due to such considerations

0
0
0
@kernellogger @gregkh i could have deleted SLAB in 6.7 😱
1
0
3

@kernellogger I was kinda hoping that it would be Kernel 6.6.6. 😏 😈

1
0
0

@torspedia

I guess many people did. 😏

0
0
0
Edited 1 year ago
@vbabka @kernellogger @gregkh
users will see the deprecation at least....
1
0
1

4/ And in case anyone wonders:

Yes, the projected EOL in December 2026 will mean that 6.6.y will be supported *three* years in total, and not *two*, which since about nearly a year is the new official expected support timeframe for longterm kernels[1].

I'm pretty sure the stable team did this to not end support for 6.6 before 6.1 becomes EOL.

Hence next years longterm kernel will likely be the first since a while that will only get two years of support.

[1] https://fosstodon.org/@corbet@social.kernel.org/111104918436086518

1
2
2

@hyeyoo @kernellogger @gregkh @vbabka F*'em obsolete users! If a retard still used SLAB in 6.5, they probably also ran it on an Itanium… πŸ˜†

0
0
2
@kernellogger @gregkh 4 kernels in a row with an EOL around the same time blobcatblep
1
0
0
@kernellogger Yes, that's why the dates are like this.

And I just realized, we never publicly said anything about "2 years support for LTS", the kernel.org page has always had longer dates, all we said was "we will not be doing 6 years anymore".

So no one actually looks at the documentation we write (i.e. the web page), AND I didn't even remember that we had written that back in February of this year, this feels like a "write once, read never" type of file...
1
1
5
@vegard @kernellogger Ah, so back in February we made it 4 years, not 2, so why did everyone freak out a month ago thinking it was going to be 2 years?

Math with dates is hard :)
1
0
1
@exi @kernellogger We will finally get to relax at that time. Or at least we can look forward to the potential to relax, providing us hope when buried neck-deep in stable backports...
0
1
3

@gregkh @kernellogger In February you set 6.1 at 4 years; I think people were scared the _next_ one (i.e. 6.6) was going to be 2 years.

1
0
0

@vegard @gregkh

maybe, but I don't think so.

I think the "Longterm is going back to 2 years per release" news only reached some people[1], but the wider public only learned about it through @corbet's talk and OSSEU[2].

[1] I shared it here, but I think it wasn't even made a proper news entry on @LWN

[2] https://fosstodon.org/@corbet@social.kernel.org/111104918436086518

0
0
0